Changes - tesseract-ocr - An OCR Engine that was developed at HP Labs between 1985 and 1995... and now at Google.

726 - 702 of 726 Older ›

Committed Changes

Rev	Commit log message	Date	Author
r726	Conform to the hocr spec: hocr doesn't have ocr_word, but instead has ocrx_word. Tested with ExactImage's hocr2pdf. $ tesseract phototest.tif phototest hocr $ hocr2pdf -i phototest.tif -o ./phototest.pdf < ./phototest.hocr $ evince phototest.pdf See: https://docs.google.com/document/preview?id=1QQnIQtvdAC_8n92-LhwPcjtAUFwBlzE8EWnKAxlgVf0	Yesterday (20 hours ago)	david.eger
r725	Provide better paragraph segmentation without having to run fully automatic layout analysis.	May 9, 2012	david.eger
r724	wiki update	Apr 26, 2012	zdenop
r723	fix issue Issue 684 : show correct line in failure message "Couldn't find a matching blob"	Apr 22, 2012	zdenop
r722	Fix Issue 678	Apr 17, 2012	zdenop
r721	Fix an issue where we sometimes leave a dangling outline->loop pointer during chopping.	Apr 16, 2012	david.eger
r720	Fix a shapetable serialization issue -- sizeof(bool) is not portable. See http://code.google.com/p/tesseract-ocr/issues/detail?id=669	Apr 16, 2012	david.eger
r719	Add some documentation on how to use config files and user dictionaries.	Apr 9, 2012	david.eger
r718	fix Mingw shared build	Apr 2, 2012	zdenop
r717	add Doxyfile (doxygen config) to distribution	Apr 2, 2012	zdenop
r716	change comments to doxygen block comments (api)	Mar 30, 2012	zdenop
r715	fix doxygen warnings	Mar 30, 2012	zdenop
r714	Missing close bracket.	Mar 28, 2012	david.eger
r713	Good moming, Good aftemoon. During our initial chopping for each word, pay attention to whether a dangerous ambiguity (like rn <-> m) would lead us to a dictionary word. If so, make sure that blob gets chopped so that we can evaluate said dictionary word during the segmentation search. Large accuracy improvement, especially on English printed books (~9%).	Mar 28, 2012	david.eger
r712	improve doxygen config; fix doxygen warnings for baseapi.h	Mar 28, 2012	zdenop
r711	Recording segmentation state for a choice at LogNewChoice() time was a bad idea -- a VIABLE_CHOICE's Blob->NumChunks can be modified as we go by a call from Dict::LogNewSplit(). Relying on the auxilury segmentation_state makes alt choices sometimes reference the wrong blobs.	Mar 28, 2012	david.eger
r710	fix 'make dist' for MinGW+MSYS	Mar 24, 2012	zdenop
r709	another fix MinGW+MSYS	Mar 24, 2012	zdenop
r708	improve autools system (mingw+msys fix); implementation of --disable-tessdata-prefix	Mar 22, 2012	zdenop
r707	Restore VC++ compatibility for paragraphs.cpp. Missed a __func__ addition in the last merge.	Mar 21, 2012	david.eger
r706	Fix Issue 645, a char signed/unsigned issue in paragraphs.cpp. When constructing our debug strings, our simple UTF-8 processing should skip all non-ASCII chars.	Mar 20, 2012	david.eger
r705	fixed build in java directory; create documentation package with 'make doc-pack'	Mar 15, 2012	zdenop
r704	fopen() should use binary mode (issue 70)	Mar 11, 2012	zdenop
r703	Fix dawg2wordlist crash on Windows caused by fopening dawg file in "r" instead of "rb" mode.	Mar 10, 2012	tomp2010
r702	+ fix VS2008 warning about "non dll-interface class tesseract::LTRResultIterator used as base for dll-interface class tesseract::ResultIterator" by making LTRResultIterator also visible. + Changed Project preprocessor definition of WINDLLNAME, because stringizing operator doesn't seem to work when initializing tessedit_module_name in ccutil/ccutil.cpp (which was omitted in previous fixes). + Update vs2008/tesshelper.py for new public header files. patch from Tom Powers (https://groups.google.com	Mar 8, 2012	zdenop

726 - 702 of 726 Older ›